186 ANNALS NEW YORK ACADEMY OF SCIENCES 



Tortola of the Virgin Islands. Its absence from the other Virgin 

 Islands is probably due to difference in the habitat conditions. It is 

 nearly allied to Anolis ricordii of Hispaniola. 



There is little variation in this series. In No. 13234 the tail crest is 

 unusually high, fully as high as in A. ricordii of Hispaniola, but the 

 scale characters which distinguish cuvi&ri from ricordii are perfectly 

 constant. 



Seven out of eight stomachs examined contained the remains of large 

 beetles; one, a large phasmid; one, remains of heteropterous bugs; and 

 one, a mass of skin of Anolis cuvieri (doubtless its own). The boys say 

 that it eats berries and fruits, and in the coffee plantations it is said to 

 eat coffee berries. It seems probable that vegetable matter forms only a 

 small proportion of its food, as in Anolis cristatellus. 



Anolis cristatellus Dumeril and Bibron 



The following localities are represented by three hundred specimens in 

 the collection: Adjuntas, Aibonito, Bayamon, Cataiio, Coamo Springs, 

 Ensenada, Maricao, Mayaguez, Salinas, San Truce, Desecheo Island, 

 Mona, Vieques, and Culebra. This species ranges everywhere in Porto 

 Bico. It occurs also on Mona and Desecheo Island, to the west, and ex- 

 tends through the entire Virgin group to the east. I do not regard the 

 records of this species from Santo Domingo as valid. A nearly allied 

 form is found in the Turk's Islands (Anolis albipalpebralis Barbour) . 



Stejneger regarded the Anolis of Mona Island as a species distinct from 

 A. cristatellus, differing in having larger scales on the head; hence fewer 

 loreals and fewer scales between the occipitals and the semicircles; in 

 having a much higher tail crest, and a somewhat peculiar coloration. I 

 cannot agree in this separation. Specimens from Ensenada and Coamo 

 Springs agree exactly with those from Mona, while A. cristatellus from 

 Culebra Island has an even higher caudal crest than those from Mona. 

 The coloration of the Mona specimens taken on limestone is not ordinarily 

 seen in Porto Rican cristatellus, but specimens taken on limestones at 

 Ensenada, Salinas, and Coamo Springs are similarly colored. Ordinary 

 cristatellus with low tail crests occur in the same area, and it is obviously 

 impossible to separate them. The species does differ somewhat on the 

 various islands, but the variation curves overlap too greatly to warrant 

 even subspecific distinction. The number of scales between the occipital 

 and the semicircles varies as follows in forty specimens respectively from 

 Vieques and Mona Islands : 



